CN113567969A - Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals - Google Patents
Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals Download PDFInfo
- Publication number
- CN113567969A CN113567969A CN202111113687.6A CN202111113687A CN113567969A CN 113567969 A CN113567969 A CN 113567969A CN 202111113687 A CN202111113687 A CN 202111113687A CN 113567969 A CN113567969 A CN 113567969A
- Authority
- CN
- China
- Prior art keywords
- underwater acoustic
- signals
- signal
- hydrophone
- underwater
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses an illegal sand dredger automatic monitoring method and system based on underwater acoustic signals, wherein a hydrophone is adopted to collect underwater acoustic data, the underwater acoustic data are transmitted to an edge computing module through a signal line of the hydrophone, an underwater acoustic signal processing module in the edge computing module carries out amplification and filtering preprocessing operation on signals collected by the hydrophone, underwater background noise is removed, underwater target voiceprint signals are amplified, the processed signals are sent to a target characteristic judging module, the target characteristic judging module extracts acoustic characteristics of the signals through a signal transformation algorithm and matches the acoustic characteristics with characteristics in a pre-established sand dredger voiceprint characteristic library, and a reminding signal is sent when the matching degree reaches a set threshold value.
Description
Technical Field
The invention relates to the technical field of automatic monitoring of ships, in particular to an illegal sand production ship automatic monitoring method and system based on underwater acoustic signals.
Background
Target identification based on acoustic features is one of the important directions of research in the field of artificial intelligence at home and abroad in the last two decades. At present, the human voice recognition technology and the animal voice recognition technology in this field are well mature and widely applied. In the aspect of judging the ship voiceprint characteristics, a mature method for judging and applying the ship characteristics by utilizing a method of testing a water pool to sample the voiceprint and establishing a voiceprint library in the process of designing and constructing the ship according to the engine and the ship type of the ship is formed. On the other hand, for the extraction and identification of acoustic features of targets on the water surface and underwater, the method is also widely applied to the aspects of sunken ship salvage, underwater vehicle tracking and the like. However, because of the complex underwater environment, these technologies are mostly used in the detection fields of oceans, reservoirs and the like with small silt content and large volume of the underwater part of the target, and are not widely applied in the environments of rivers with rapid water flow and high silt content. Currently, the most common method for detecting and identifying a sand dredger is to monitor the behavior characteristics of the sand dredger, a VTS monitoring station is formed by combining shore-based radar, video monitoring, very high frequency communication or AIS data, and whether a target is the sand dredger or not is comprehensively judged through the target behavior characteristics and ship registration information.
The existing underwater interception technology and VTS monitoring technology mainly have the following problems: firstly, current interception technique under water can't distinguish, filter adopting sand ship, inland river environment background sound, can't effectively establish the clear adopting sand ship acoustics characteristic collection environment of background, is difficult to the practicality in practical application. Secondly, a voiceprint library based on the characteristics of the ship engine is established, the characteristics of the sand dredger engines with different powers and different installation positions need to be considered, the illegal sand dredger is mostly modified illegally, and the voiceprint library covering all ship types cannot be established depending on different ship types. Thirdly, the VTS monitoring technology mainly combines radar echo and AIS non-registration information to judge suspected sand production ship information, but at night, the ship has behaviors of anchoring, rest, slow running and the like and can be confused with sand production behaviors. Meanwhile, the opening rate of the AIS equipment in inland rivers is generally less than 2, so the number of ships which can participate in the judgment of the AIS is not enough to support the application.
Disclosure of Invention
The invention aims to provide an illegal sand dredger automatic monitoring method and system based on underwater acoustic signals according to underwater acoustic characteristics of a hydrophone monitoring inland river aiming at the defects in the prior art.
The invention discloses an illegal sand dredger automatic monitoring method based on underwater acoustic signals, which is characterized in that a hydrophone is adopted to collect underwater acoustic data, the underwater acoustic data are transmitted to an edge computing module through a signal line of the hydrophone, an underwater acoustic signal processing module in the edge computing module performs amplification and filtering preprocessing operations on signals collected by the hydrophone, underwater background noise is removed, underwater target voiceprint signals are amplified, the processed signals are sent to a target characteristic judging module, the target characteristic module extracts acoustic characteristics of the signals through a signal transformation algorithm and matches the acoustic characteristics with characteristics in a pre-established sand dredger voiceprint characteristic library, and when the matching degree reaches a set threshold value, a reminding signal is sent.
Furthermore, the underwater acoustic signal processing module sets sampling receiving parameters, receives signals collected by the hydrophone, performs time-frequency domain transformation on the sampling data by using an FFT algorithm according to the set parameters, eliminates high-frequency signals after the transformation is completed, and retains low-frequency signals, wherein an FFT change formula is as follows:
wherein, X is discrete sound signal, W is periodic phase of sound wave, k is periodic sequence, and N is sampling period number.
Further, the underwater acoustic signal processing module extracts the retained low-frequency signal, and removes the background noise of the low-frequency signal by using an LMS adaptive noise reduction algorithm, wherein the LMS filtering algorithm process is as follows:
the method comprises the following steps: given w (0), and 1< μ <1/λ max;
step two: calculating an output value: y (k) = w (k) tx (k);
step three: calculating an estimation error: e (k) = d (k) -y (k);
step four: and (3) updating the weight: w (k +1) = w (k) + μ e (k) x (k);
wherein w is an input sound signal, y is an output signal, Tx is a filtering parameter, d is an expected signal value, μ is a characteristic root value of the gradient, and λ max is a characteristic root value of the maximum gradient matrix;
further, the target feature determination module performs a continuous spectrum transformation algorithm on the filtered signal, so as to reduce the difference caused by the high and low sound volume and the phase shift of the sound wave, and performs energy cepstrum on the frequency domain spectrum, wherein the cepstrum formula is as follows:
wherein, M (f) is a signal value after cepstrum, and a, b and c are conversion coefficients;
normalizing the cepstrum signals, normalizing sound waves in different periods into a unified calculation frame, aligning fundamental tones and 0-hour positions, and recursively calculating the offset of each frequency band and time sequence by using a least square method according to a calculation formula as follows:
wherein x is the normalized sound signal value, f1Is a cepstrum signal value, N is a signal sampling number, and t is a conversion parameter;
and extracting full-band voiceprint continuous spectrums and energy density statistical spectrums of sub-bands, and calculating different energy characteristic parameters according to the statistical spectrums to form a judgment parameter data packet.
Further, a sand dredger voiceprint feature library is established, test field and test sand dredger data are collected through a hydrophone, an underwater acoustic signal processing module carries out amplification and filtering preprocessing operation on signals collected by the hydrophone, a target feature judgment module extracts acoustic feature parameters of the signals through a signal transformation algorithm, and by means of multilayer convolution and recursive analysis of the feature parameters, invalid parameters are eliminated, valid parameters are reserved, and a preliminary model library is formed, and the method comprises the following steps:
the method comprises the following steps: dividing a sound sample of the sand dredger into a plurality of frequency bands to form sample sets of different frequency bands;
step two: dividing a sound sample of the sand dredger into a plurality of segments according to sampling time to form a sample set of different segments;
step three: inputting the sample sets of different frequency bands and segments into a convolution network according to frequency spectrum and time sequence, and extracting characteristics;
step four: connecting samples with different time sequences and the same frequency band to form a multilayer connection network;
step five: and identifying and sorting the model library, and performing convolution calculation to improve the accuracy of the model and form a final judgment model.
The invention also discloses an illegal sand dredger automatic monitoring system based on the underwater acoustic signal, which adopts the illegal sand dredger automatic monitoring method based on the underwater acoustic signal, and the system comprises a hydrophone, an underwater acoustic signal processing module and a target characteristic judging module, wherein the hydrophone is arranged under the water, the underwater acoustic signal processing module and the target characteristic judging module are software modules and are arranged in an edge calculating module, and the edge calculating module is arranged in a monitoring tower pole fixed on the bank;
the connection mode is as follows: the hydrophone is connected with the edge calculation module underwater sound signal processing module through a signal cable, and is connected with the shore-based monitoring tower through a power cable;
the system formed by the connection mode has the following monitoring modes to the sand dredger: the hydrophone receives underwater acoustic signals, transmits the underwater acoustic signals to the underwater acoustic signal processing module in the edge computing module through the signal cable, and is processed by the underwater acoustic signal processing module and then delivered to the target characteristic judging module for target characteristic identification and judgment.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention relates to a technical scheme for capturing the voiceprint characteristics of a sand dredger during working in a noisy and turbid underwater environment of an inland river based on limited test data and separating the voiceprint characteristics from a background environment. The voiceprint characteristics are collected by a hydrophone, preliminary electric signal noise processing is carried out, and the voiceprint data is subjected to noise reduction, frequency spectrum conversion and characteristic extraction by the underwater acoustic signal processing module;
(2) the invention utilizes the data processed by the underwater acoustic signal processing module, combines with a sand production ship characteristic database to continuously monitor the sand production behavior characteristics of inland rivers under different working conditions and environments and judge the numerical calibration of the sand production behavior conformity, wherein, the sand production ship characteristic database can be based on the acquisition by utilizing a hydrophone under different environments, and is calibrated by manually combining with a machine learning algorithm after being processed by the underwater acoustic signal processing module, and the sand production ship voiceprint characteristics are extracted. The sand production ship numerical judgment is based on the voiceprint data processed by the sand production ship characteristic database and the sand production ship processing algorithm, and the standardized numerical classification system formulated according to the invention is used for classifying the signal activity intensity to form sand production activity intensity classifications of different levels.
Drawings
FIG. 1 is a diagram of the system components disclosed herein;
FIG. 2 is a flow chart of the disclosed method;
FIG. 3 is an algorithm framework for hydrophone feature extraction and transformation as disclosed in the present invention.
Detailed Description
As shown in fig. 1, the invention discloses an illegal sand dredger automatic monitoring system based on underwater acoustic signals, which comprises a hydrophone 1, an underwater acoustic signal processing module 2 and a target characteristic judgment module 3, wherein the hydrophone 1 is arranged under water, the underwater acoustic signal processing module 2 and the target characteristic judgment module 3 are software modules and are installed in an edge calculation module 4, and the edge calculation module 4 is arranged in a monitoring tower pole fixed on the shore.
The connection mode is as follows: the hydrophone 1 is connected with the edge calculation module 4 through a signal cable, and the hydrophone 1 is connected with a shore-based monitoring tower through a power cable.
The system formed by the connection mode has the following monitoring modes to the sand dredger: the hydrophone 1 receives underwater acoustic signals and transmits the underwater acoustic signals to the underwater acoustic signal processing module 2 in the edge computing module 4 through a signal cable, and the underwater acoustic signals are processed by the underwater acoustic signal processing module 2 and then delivered to the target characteristic judging module 3 for target characteristic identification and judgment.
In this embodiment, the edge calculation module 4 is arranged at a fixed point on the shore, the hydrophone 1 is connected with the edge calculation module 4 by adopting a cable and a signal cable, and is arranged at the bottom of a river several meters away from the shore, and after the power-on and the normal transmission of signals are confirmed, normal monitoring activities can be started; wherein, the hydrophone 1 with the sensitivity higher than-220 hz is adopted to collect the underwater acoustic data.
As shown in fig. 2 and 3, the automatic monitoring method of the illegal sand production ship automatic monitoring system based on the underwater acoustic signal disclosed by the invention comprises the following steps:
the method comprises the following steps: the hydrophone 1 monitors the underwater sound characteristics of the suspicious vessel 5 and transmits the underwater sound signals to the edge calculation module 4 through the signal cable.
Step two: an underwater acoustic signal processing module 2 in the edge calculation module receives the underwater acoustic signal, performs amplification and filtering preprocessing operations on the underwater acoustic signal, and transmits the filtered signal to a target characteristic judgment module 3.
The data processing method of the underwater sound signal processing module 2 comprises the following steps:
step 201: the underwater sound signal processing module receives acoustic data for filtering: and setting sampling receiving parameters and receiving hydrophone data. And according to the pre-set parameters, carrying out time-frequency domain transformation on the sampled data by using an FFT algorithm, after the transformation is finished, removing high-frequency signals and reserving low-frequency signals. The FFT variation formula is as follows:
wherein, X is discrete sound signal, W is periodic phase of sound wave, k is periodic sequence, and N is sampling period number.
Step 202: extracting a reserved signal, and removing background noise of the reserved signal by using an LMS adaptive noise reduction algorithm to reduce the influence of echo and clutter on the signal and improve the signal quality, wherein the LMS filtering algorithm process comprises the following steps:
(1) given w (0), and 1< μ <1/λ max;
(2) calculating an output value: y (k) = w (k) tx (k);
(3) calculating an estimation error: e (k) = d (k) -y (k);
(4) and (3) updating the weight: w (k +1) = w (k) + μ e (k) x (k).
Wherein w is an input sound signal, y is an output signal, Tx is a filtering parameter, d is an expected signal value, mu is a characteristic root value of the gradient, λ max is a characteristic root value of a maximum gradient matrix, and w (0) is a first input sound signal of the array;
in the above formula, the w (0) and e (k) limit values are obtained according to data collected in the earlier quiet environment for underwater acoustic environment and ship driving drag saw tests. The extracted signals eliminate random noise which greatly influences the judgment of target characteristics, and keep normal background noise of underwater environment, ship engines and comprehensive sound veins of sand mining working conditions.
Step three: and the target characteristic judgment module 3 receives the filtered underwater sound signal, extracts the characteristics of the underwater sound signal, puts the characteristics into a characteristic library, matches the extracted characteristics with the characteristics in the characteristic library, and gives an alarm if the sand dredger is the sand dredger after the matching degree reaches a set threshold value. The method specifically comprises the following steps:
step 301: the target characteristic judgment module performs a continuous spectrum transformation algorithm on the filtered signals, so that differences caused by high and low sound volume and sound wave phase displacement are reduced, sound waves in different time periods are normalized in a unified calculation frame, and subsequent calculation is facilitated. The spectrum transformation steps are as follows:
(1) performing energy cepstrum on the frequency domain spectrum, wherein a cepstrum formula is as follows:
wherein, M (f) is the signal value after cepstrum, and a, b, c are Mel cepstrum coefficients.
And selecting the coefficient of the spectral line with the minimum error and the optimal result from the cepstrum results of the multiple samples according to a plurality of previous tests.
(2) Normalizing the cepstrum signal, aligning fundamental tone and 0 time position, and recursively calculating the offset of each frequency band and time sequence by using a least square method, wherein the least square method has the following calculation formula:
wherein x is the normalized sound signal value, f1Is a cepstrum signal value, N is a signal sampling number, and t is a conversion parameter;
step 302: and extracting full-band voiceprint continuous spectrums and energy density statistical spectrums of sub-bands of the acquired signals, and calculating energy characteristic parameters such as peak values, valley values, slopes and accumulated energy in different frequency bands according to the full-band voiceprint continuous spectrums and the energy density statistical spectrums, so as to form a judgment parameter data packet.
Step 303: and (3) carrying out joint judgment on the judgment parameters and a pre-established joint characteristic model library comprising an underwater sound background model library, an underwater ship power model library, an underwater noise model library and a sand mining behavior characteristic library, finding out the characteristics with highest similarity and highest conformity, and giving out the characteristic conformity percentage according to the effectiveness of the characteristic factors.
Step 304: the underwater sound signals in different time periods are continuously monitored, judged and retained, secondary judgment is carried out according to sound continuity and peak frequency, secondary judgment is carried out on a time domain spectrum, false alarm data are eliminated, signals which continuously have certain characteristics are provided, and weighting is carried out to form four indexes of sand mining activity intensity, background noise intensity, underwater sound volume and surrounding ship activity intensity, when the value of each index is higher than 6, the index is high in reliability and strength.
In the process, the pre-established acoustic print feature library of the sand dredger is established by the following method:
(1) selecting a test field, collecting short-time and long-time underwater acoustic signals of multiple batches, recording the collection time, and collecting the passage condition of the ships on the airway at the time interval;
(2) selecting a test ship, and acquiring vocal print characteristics under different distances, different working conditions and different loads by using a hydrophone;
(3) extracting characteristic parameters according to the steps two to three of the technical scheme of the method;
(4) and performing recursive analysis on the characteristic parameters by utilizing multilayer convolution, eliminating invalid parameters, and reserving valid parameters to form a primary model library. The multilayer convolution is established as follows:
(401) dividing the sound sample into a plurality of frequency bands to form a sample set of different frequency bands;
(402) dividing the sound sample into a plurality of segments according to sampling time to form a sample set of different segments;
(403) inputting the sample sets of different frequency bands and segments into a convolution network according to frequency spectrum and time sequence, and extracting characteristics;
(404) connecting samples with different time sequences and the same frequency band to form a multilayer connection network;
(5) identifying and sorting the model library, performing convolution calculation, improving the accuracy of the model, and forming a final judgment model, wherein the identification technology is implemented in the following process:
(501) checking the characteristic line difference and the energy difference between the sample library and the underwater sound background, and checking the characteristic line difference and the energy difference between the sample libraries;
(502) checking the noise level and the classification aggregation rate of the sample library;
(503) removing samples with the height consistent with that of the background characteristic line and samples with high noise and low polymerization rate;
(504) and running the test data, checking the effectiveness of the corrected sample, and if the false alarm rate and the false alarm rate of the test data are lower than expected, determining that the sample is effective.
Claims (6)
1. An illegal sand dredger automatic monitoring method based on underwater acoustic signals is characterized in that: the method comprises the steps that a hydrophone (1) is used for collecting underwater acoustic data, the underwater acoustic data are transmitted to an edge computing module (4) through a signal line of the hydrophone, an underwater acoustic signal processing module (2) in the edge computing module carries out amplification and filtering preprocessing on signals collected by the hydrophone (1), underwater background noise is removed, underwater target voiceprint signals are amplified, the processed signals are sent to a target characteristic judging module (3), the target characteristic judging module (3) extracts acoustic characteristics of the signals through a signal transformation algorithm and matches the acoustic characteristics with characteristics in a pre-established sand dredger voiceprint characteristic library, and when the matching degree reaches a set threshold value, a reminding signal is sent.
2. The illegal sand dredger automatic monitoring method based on underwater acoustic signals according to claim 1, characterized in that: the underwater acoustic signal processing module (2) sets sampling receiving parameters, receives signals collected by the hydrophone (1), performs time-frequency domain transformation on the sampling data by using an FFT algorithm according to the set parameters, eliminates high-frequency signals after the transformation is completed, and retains low-frequency signals, wherein an FFT change formula is as follows:
wherein, X is discrete sound signal, W is periodic phase of sound wave, k is periodic sequence, and N is sampling period number.
3. The illegal sand dredger automatic monitoring method based on underwater acoustic signals according to claim 2, characterized in that: the underwater acoustic signal processing module (2) extracts the reserved low-frequency signal, and eliminates the background noise of the low-frequency signal by using an LMS adaptive noise reduction algorithm, wherein the LMS filtering algorithm comprises the following steps:
the method comprises the following steps: given w (0), and 1< μ <1/λ max;
step two: calculating an output value: y (k) = w (k) tx (k);
step three: calculating an estimation error: e (k) = d (k) -y (k);
step four: and (3) updating the weight: w (k +1) = w (k) + μ e (k) x (k);
where w is the input sound signal, y is the output signal, Tx is the filter parameter, d is the desired signal value, μ is the characteristic root of the gradient, and λ max is the characteristic root of the maximum gradient matrix.
4. The illegal sand dredger automatic monitoring method based on underwater acoustic signals according to claim 1, characterized in that: the target characteristic judgment module (3) performs a continuous spectrum transformation algorithm on the filtered signals, reduces the difference caused by high and low sound volume and sound wave phase displacement, performs energy cepstrum on the frequency domain spectrum, and the cepstrum formula is as follows:
wherein, M (f) is a signal value after cepstrum, a, b and c are conversion coefficients, and f is an original sound signal value;
normalizing the cepstrum signals, normalizing sound waves in different periods into a unified calculation frame, aligning fundamental tones and 0-hour positions, and recursively calculating the offset of each frequency band and time sequence by using a least square method according to a calculation formula as follows:
wherein x is the normalized sound signal value, f1Is a cepstrum signal value, N is a signal sampling number, and t is a conversion parameter;
and extracting full-band voiceprint continuous spectrums and energy density statistical spectrums of sub-bands, and calculating different energy characteristic parameters according to the statistical spectrums to form a judgment parameter data packet.
5. The illegal sand dredger automatic monitoring method based on underwater acoustic signals according to any one of claims 1-4, characterized in that: the method comprises the following steps of establishing a sand dredger voiceprint feature library, collecting test field and test sand dredger data through a hydrophone (1), amplifying and filtering preprocessing operations are carried out on signals collected by the hydrophone (1) through an underwater acoustic signal processing module, extracting acoustic feature parameters of the signals through a signal transformation algorithm by a target feature judgment module (3), and eliminating invalid parameters and keeping valid parameters by utilizing multilayer convolution and recursive analysis feature parameters to form a preliminary model library, wherein the preliminary model library comprises the following steps:
the method comprises the following steps: dividing a sound sample of the sand dredger into a plurality of frequency bands to form sample sets of different frequency bands;
step two: dividing a sound sample of the sand dredger into a plurality of segments according to sampling time to form a sample set of different segments;
step three: inputting the sample sets of different frequency bands and segments into a convolution network according to frequency spectrum and time sequence, and extracting characteristics;
step four: connecting samples with different time sequences and the same frequency band to form a multilayer connection network;
step five: and identifying and sorting the model library, and performing convolution calculation to improve the accuracy of the model and form a final judgment model.
6. An illegal sand dredger automatic monitoring system based on underwater acoustic signals adopts the illegal sand dredger automatic monitoring method based on the underwater acoustic signals, which is characterized in that: the underwater acoustic monitoring system comprises a hydrophone (1), an underwater acoustic signal processing module (2) and a target characteristic judgment module (3), wherein the hydrophone (1) is arranged under water, the underwater acoustic signal processing module (2) and the target characteristic judgment module (3) are software modules and are installed in an edge calculation module (4), and the edge calculation module (4) is arranged in a monitoring tower pole fixed on the shore; the hydrophone (1) is connected with the edge calculation module (4) and the underwater sound signal processing module (2) through signal cables, and the hydrophone (1) is connected with the shore-based monitoring tower through a power cable; the hydrophone (1) receives underwater acoustic signals and transmits the underwater acoustic signals to the underwater acoustic signal processing module (2) in the edge computing module (4) through a signal cable, and the underwater acoustic signals are processed by the underwater acoustic signal processing module (2) and then delivered to the target characteristic judging module (3) for target characteristic identification and judgment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111113687.6A CN113567969B (en) | 2021-09-23 | 2021-09-23 | Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111113687.6A CN113567969B (en) | 2021-09-23 | 2021-09-23 | Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113567969A true CN113567969A (en) | 2021-10-29 |
| CN113567969B CN113567969B (en) | 2021-12-17 |
Family
ID=78174043
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111113687.6A Active CN113567969B (en) | 2021-09-23 | 2021-09-23 | Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113567969B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114460566A (en) * | 2022-01-04 | 2022-05-10 | 中国科学院声学研究所东海研究站 | Underwater sound target real-time identification method, system and device |
| CN114455042A (en) * | 2022-01-26 | 2022-05-10 | 浙江咸临智能科技有限责任公司 | Intelligent underwater sound detection system based on underwater glider |
| CN115266914A (en) * | 2022-03-28 | 2022-11-01 | 华南理工大学 | Pile sinking quality monitoring system and monitoring method based on acoustic signal processing |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1758332A (en) * | 2005-10-31 | 2006-04-12 | 浙江大学 | Speaker recognition method based on MFCC linear emotion compensation |
| CN108564956A (en) * | 2018-03-26 | 2018-09-21 | 京北方信息技术股份有限公司 | A kind of method for recognizing sound-groove and device, server, storage medium |
| CN109186752A (en) * | 2018-09-22 | 2019-01-11 | 天津大学 | Underwater sound signal acquisition, transmission and detection system based on graphics processor |
| CN109597342A (en) * | 2019-01-16 | 2019-04-09 | 郑州轻工业学院 | A kind of the sand dredger monitoring device and method of dynamic group net INTELLIGENT IDENTIFICATION |
| CN109884591A (en) * | 2019-02-25 | 2019-06-14 | 南京理工大学 | A kind of multi-rotor unmanned aerial vehicle acoustical signal Enhancement Method based on microphone array |
| CN111489763A (en) * | 2020-04-13 | 2020-08-04 | 武汉大学 | Adaptive method for speaker recognition in complex environment based on GMM model |
| CN112684750A (en) * | 2020-12-04 | 2021-04-20 | 江苏科技大学 | Underwater intelligent monitoring equipment and method for sand dredger |
-
2021
- 2021-09-23 CN CN202111113687.6A patent/CN113567969B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1758332A (en) * | 2005-10-31 | 2006-04-12 | 浙江大学 | Speaker recognition method based on MFCC linear emotion compensation |
| CN108564956A (en) * | 2018-03-26 | 2018-09-21 | 京北方信息技术股份有限公司 | A kind of method for recognizing sound-groove and device, server, storage medium |
| CN109186752A (en) * | 2018-09-22 | 2019-01-11 | 天津大学 | Underwater sound signal acquisition, transmission and detection system based on graphics processor |
| CN109597342A (en) * | 2019-01-16 | 2019-04-09 | 郑州轻工业学院 | A kind of the sand dredger monitoring device and method of dynamic group net INTELLIGENT IDENTIFICATION |
| CN109884591A (en) * | 2019-02-25 | 2019-06-14 | 南京理工大学 | A kind of multi-rotor unmanned aerial vehicle acoustical signal Enhancement Method based on microphone array |
| CN111489763A (en) * | 2020-04-13 | 2020-08-04 | 武汉大学 | Adaptive method for speaker recognition in complex environment based on GMM model |
| CN112684750A (en) * | 2020-12-04 | 2021-04-20 | 江苏科技大学 | Underwater intelligent monitoring equipment and method for sand dredger |
Non-Patent Citations (2)
| Title |
|---|
| 郑长松 等: "《装甲车辆故障诊断技术》", 31 May 2019, 北京理工大学出版社 * |
| 陈宗海: "《***仿真技术及其应用 第7卷》", 31 August 2005, 中国科学技术大学出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114460566A (en) * | 2022-01-04 | 2022-05-10 | 中国科学院声学研究所东海研究站 | Underwater sound target real-time identification method, system and device |
| CN114460566B (en) * | 2022-01-04 | 2024-04-30 | 中国科学院声学研究所东海研究站 | Underwater sound target real-time identification method, system and device |
| CN114455042A (en) * | 2022-01-26 | 2022-05-10 | 浙江咸临智能科技有限责任公司 | Intelligent underwater sound detection system based on underwater glider |
| CN114455042B (en) * | 2022-01-26 | 2022-10-11 | 浙江咸临智能科技有限责任公司 | Intelligent underwater sound detection system based on underwater glider |
| CN115266914A (en) * | 2022-03-28 | 2022-11-01 | 华南理工大学 | Pile sinking quality monitoring system and monitoring method based on acoustic signal processing |
| CN115266914B (en) * | 2022-03-28 | 2024-03-29 | 华南理工大学 | Pile sinking quality monitoring system and method based on acoustic signal processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113567969B (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113567969B (en) | Illegal sand dredger automatic monitoring method and system based on underwater acoustic signals | |
| CN110135316B (en) | Automatic detection and extraction method for low-frequency line spectrum in ship radiation noise | |
| CN111414832B (en) | Real-time online recognition and classification system based on whale dolphin low-frequency underwater acoustic signals | |
| CN106052852B (en) | A kind of detection method and device of pulse acoustical signal | |
| Wang et al. | ia-PNCC: Noise Processing Method for Underwater Target Recognition Convolutional Neural Network. | |
| CN105823492B (en) | Weak target signal extracting method in a kind of interference of ocean current | |
| CN111175729A (en) | Real-time online monitoring and early warning system based on whale high-frequency sonar signals | |
| CN109741759B (en) | Acoustic automatic detection method for specific bird species | |
| CN113077813B (en) | Ship noise identification method based on holographic spectrum and deep learning | |
| Sorensen et al. | Passive acoustic sensing for detection of small vessels | |
| CN111914641B (en) | Target depth identification method and system based on modal intensity matching analysis | |
| CN114513262B (en) | Underwater acoustic signal processing method based on underwater glider | |
| CN108281155A (en) | Application of the zero passage detection method based on rayleigh scattering in optical fiber water listens system | |
| CN116878755A (en) | Method for monitoring submarine gas leakage based on bubble passive acoustics | |
| CN114927141B (en) | Method and system for detecting abnormal underwater acoustic signals | |
| Qiao et al. | Spectral entropy based dolphin whistle detection algorithm and its possible application for biologically inspired communication | |
| CN106382978A (en) | Automatic target detection method for submarine sound signal generated through time-varying ship excitation | |
| CN115932808A (en) | Passive sonar intelligent detection method based on multi-feature fusion | |
| Sánchez-García et al. | A novel image-processing based method for the automatic detection, extraction and characterization of marine mammal tonal calls | |
| CN115424639B (en) | Dolphin sound endpoint detection method under environment noise based on time-frequency characteristics | |
| Urazghildiiev et al. | Acoustic detection and recognition of fin whale and North Atlantic right whale sounds | |
| CN116405127B (en) | Compression method and device of underwater acoustic communication preamble signal detection model | |
| CN112834577A (en) | Correlation method for judging marine environment magnetic field and seawater conductivity | |
| Bougher et al. | Generalized marine mammal detection based on improved band-limited processing | |
| CN112185396B (en) | Offshore wind farm biological monitoring method and system based on passive acoustics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |